Sparse and Robust Reject Option Classifier using Successive Linear Programming

02/12/2018

∙

by   Kulin Shah, et al.

∙

0

∙

share

In this paper, we propose a new sparse and robust reject option classifier based on minimization of l_1 regularized risk under double ramp loss L_dr,ρ. We use DC programming to find the risk minimizer. The algorithm solves a sequence of linear programs to learn the reject option classifier. Moreover, we show that the risk under L_dr,ρ is minimized by generalized Bayes classifier in the reject option setting. We also provide the excess risk bound for L_dr,ρ. We show the effectiveness of the proposed approach by experimenting it on several real world datasets.